Primary HIV infection appears to be effectively controlled by both CTL and antibody responses. The kinetics and peak level of CTL response correlates with a reduction in the levels of viremia in HIV-infected individuals. This protective role of CTL is not sustained during progression to AIDS. Immune escape has been proposed to be one of the mechanisms for the failure of anti-HIV CTL. In the early CTL population, there is a preponderance of responses to Envelope, Gag (structural) and Nef (regulatory) proteins of HIV-1. Gag and Nef are made in the cytosol before delivery to the plasma membrane by post-translational modification by N-myristoylation. It is therefore conceivable that Gag and Nef are exposed to the cytosolic ubiquitin-proteosome pathway, which generates epitopes for presentation by HLA class I antigens. However, the generation of env-specific CTL is hard to conceptualize as the envelope glycoprotein is translocated into the ER, the compartment where antigenic peptide epitopes assemble with class I molecules. This mode of Env biosynthesis in the secretory pathway is not optimal for efficient generation and presentation of peptides to HLA class I. However, HIV infected individuals do generate CTL that are envelope specific indicating that HIV Env is accessible to the antigen presentation pathway during its biosynthesis. It is likely that the exposure of Env to the proteosome is only limited and thereby would generate only a limited set of CTL epitopes. The applicant hypothesize that specific ER degradation of HIV Env has the potential to generate antigenic peptides more efficiently in the cell and thereby expand the repertoire of antigen selection and presentation. To achieve a sequence-specific selective degradation of HIV envelope glycoproteins in the ER, the applicant will use the HIV-1 Vpu protein. Previous in vitro studies have shown that HIV-1 Vpu induces the ER degradation of HIV-1 envelope glycoproteins appended to CD4 trans-membrane and cytoplasmic domains. We hypothesize that this property of the HIV Vpu protein could enhance the generation of Env peptides for presentation by MHC class I antigens in vivo. Since Env-CD4 proteins are anchored to the cell membrane, they would also retain the ability to induce antibody responses with or without Vpu. The applicant will test this notion by introducing in animals (mice and monkeys) the genes encoding EnvCD4 and Env-Gag-CD4 fusion proteins as DNA vaccines in the presence and absence of the Vpu expression plasmid. The applicant will perform initial experiments in mice to test the proof of principle that specific ER degradation of Env would enhance env-specific CTL and antibody responses. Prospective vaccine constructs will be used to immunize rhesus macaques and immune protection will be assessed by challenging vaccinated monkeys with pathogenic SHIV isolates. This line of inquiry has the potential to reveal the modality of antigen presentation for the design of an effective AIDS vaccine.